Date: Wed, 29 Aug 2001 17:32:09 -0700 From: Sean Chittenden <sean@chittenden.org> To: Kory Hamzeh <kory@avatar.com> Cc: Trevin Chow <trevin@mail.com>, questions@FreeBSD.ORG Subject: Re: Performance tuning results Message-ID: <20010829173209.A3769@rand.tgd.net> In-Reply-To: <00c301c130e9$c96f1ca0$14ce21c7@avatar.com>; from "kory@avatar.com" on Wed, Aug 29, 2001 at = 05:22:00PM References: <BCCBC5E8BA377141B125D2C5F91ABA5B032C3EEB@red-msg-07.redmond.corp.microsoft.com> <00c301c130e9$c96f1ca0$14ce21c7@avatar.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] > The Benchmark: I have a shell script that converts about 650 JPG files with > an average size of 30K to JPG thumbnails with an average size of 8K. I > rebooted the system before each test and entered the exact same commands. I > used the "time" command and just recorded the total elapsed time. A better benchmark would be to create a ton of itty bitty files like from a mail server... the reason there isn't much time diff between the two is because you picked a pretty CPU intensive task to benchmark. Try something w/ more files, that are created more requently. Toss together a quick program that forks off 20 copies of itself that write 500byte files full of zeros and then deletes them... you should see a pretty decent sized difference. -sc -- Sean Chittenden [-- Attachment #2 --] -----BEGIN PGP SIGNATURE----- Comment: Sean Chittenden <sean@chittenden.org> iEYEARECAAYFAjuNiYcACgkQn09c7x7d+q0PbgCaAqlV8gPw1IRYDi1O9so+/XH1 6DkAn0y+/B+apm7lmG9zJPcrd8jaPCpm =i5vC -----END PGP SIGNATURE-----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010829173209.A3769>